Botanical and entomological aspects of the pollination biology of macadamia and cashew were studied. Flowers of both species require insect visitation for high yields. The level of bee visitation to macadamia flowers is correlated with nut set. The flowers on a raceme open over a period of 6 to 12 days. The duration of attractiveness of individual flowers varied from 17 to 109 hours. A peak of flower opening occurs in early to mid afternoon. Nectar secretion is greater at night than during the day. If the nectar is not removed during the day, approximately 50% of it is resorbed or evaporated. In open flowers nearly all the nectar is removed by visiting insects. Macadamia nectar is sucrose dominant.
Fifty-nine species of insects and birds visited macadamia flowers but only two species Apis mellifera and Trigona carbonaria, were common. Apis mellifera was present at all of 15 widely dispersed sites but T. carbonaria was present at only 10 of the sites. Population size of Trigona, but not honey bees, was significantly positively correlated with extent of surrounding eucalypt vegetation. Both species foraged on both heavily and lightly flowering trees and on both inner shaded racemes and outer sunny ones, but preferred outer ones. Trigona foraged for only 7 hours of each day compared to 10 hours for honey bees. Trigona reached a foraging peak earlier in the day. Both bee species were present for the major period of macadamia flowering.
Observations of Trigona and honey bee foraging on macadamia racemes suggest that both species, but particularly Trigona, contribute to pollination. Trigona mainly collected pollen and this activity resulted in intimate contact with the stigma. Honey bees mainly collected nectar and made less contact with the stigma. Both honey bees and Trigona responded to racemes rich in pollen and nectar by remaining longer at those racemes and visiting more flowers on them. Honey bees work flowers more quickly, visiting more flowers in a given time. Racemes which were enclosed by cages allowing visitation only by Trigona yielded a nut set equal to open pollinated racemes showing Trigona to be efficient pollinators.
A method of transferring colonies of T. carbonaria into boxes and propagating the resulting hives was developed, increasing the potential of these bees for managed crop pollination. Approximately 50% of both honey bees and Trigona from their respective hives placed in macadamia orchards foraged for pollen. Of these pollen gatherers, Trigona had a higher preference than honey bees for macadamia flowers.
Temperature and light were the most important variables affecting flight activity of colonies of T. carbonaria. They impose thresholds to flight activity and influence the intensity of activity above the thresholds. Relative humidity, vapour pressure, cloud cover and wind speed, within the range observed, had no significant effect. A variable measuring the hours from daily peak of activity was significantly correlated to flight activity indicating an intrinsic diel pattern of activity. Daily variation in flight activity, representing an assembly of unmeasured variables, was significant.
A mean total of 442.9 flowers was produced per cashew panicle of which 32.1 (7.3%) were hermaphrodite, the remainder being male. There were statistically significant differences between breeding lines in the percentage of hermaphrodite flowers. There was no significant correlation between the number of hermaphrodite flowers and the fruit set. The absence of either male or hermaphrodite flowers at any stage of flowering would not limit fruit set. The mean initial fruit set was 5.0 per panicle, representing a percentage initial fruit set of 15.5%. The natural level of pollination is 98% suggesting that pollination is not limiting fruit set under the conditions of this experiment. Flowers are self-as well as cross-compatible. The hexose-rich nectar is secreted in small volumes.
Diversity of insect flower visitors to cashew flowers was great but only a few species were common. The only insects able to feed on floral nectar through the corolla tube were those with mouthparts longer than the depth of that tube. Of the common flower visitors, several did not make contact with the stigma when visiting flowers and were assumed to be inefficient pollinators. Three species of potentially efficient pollinators were tested for their efficiency found to be high for Apis mellifera and Ligyra sp. Both these taxa deposited large numbers of pollen grains on stigmas and had only a moderate detrimental effect on the germinability of the pollen. Wind and night flying insects played at most a minor role in pollination.